All Stories
Using approximate nearest neighbor search in real world applications
From text search and recommendation to ads and online dating, ANN search rarely works in isolation.
Photo by Donald Giannatti on Unsplash
Vespa Product Updates, December 2020
Advances in Vespa features and performance include improved tensor ranking performance, Apache ZooKeeper integration, Vespa Python API for researchers and ONNX integration.
Stateful model serving: how we accelerate inference using ONNX Runtime
There's a difference between stateless and stateful model serving.
Photo by Alice Dietrich on Unsplash
Fine-tuning a BERT model for search applications
How to ensure training and serving encoding compatibility.
From research to production: scaling a state-of-the-art machine learning system
How we implemented a production-ready question-answering application and reduced response time by more than two orders of magnitude.
Photo by Samule Sun on Unsplash
Fine-tuning a BERT model with transformers
Setup a custom Dataset, fine-tune BERT with Transformers Trainer and export the model via ONNX.
Photo by Ilya Pavlov on Unsplash
Vespa Product Updates, October 2020
Improvement to Vespa feeding APIs
Photo by ThisisEngineering RAEng on Unsplash
Vespa Product Updates, September 2020
Introducing ONNX-Runtime, Hamming Distance Metric, Conditional Update Performance Improvements and Compressed Transaction Log with Synced Ack
Efficient open-domain question-answering on Vespa.ai
In this post, we reproduce the state-of-the-art baseline for retrieval-based question-answering systems within a single, scalable production ready application on Vespa.ai.
Vespa Product Updates, August 2020
Introducing NLP with Transformers, Grafana how-to, Improved GEO Search Support, Query Profile Variants Optimizations, & Build on Debian 10
Vespa Product Updates, June 2020
Announcing support for approximate nearest neighbor vector search which can be combined with filters and text search with state-of-the art performance
Never miss a story from us, subscribe to our newsletter